Port Reference

Search 5,985 TCP and UDP ports by number, service or category. Get the protocol, what runs there, RFC references and whether it is safe to expose.

Advertisement

Look Up Any TCP or UDP Port Number

This port reference searches a database of 5,985 registered TCP and UDP ports and tells you what each one is for, which protocol it uses, what category of service it belongs to, and what the security implications of exposing it are. Type a port number to identify it, or type a service name to find the port it runs on — the search works in both directions.

It is the lookup you reach for when a firewall log shows traffic on a port you do not recognise, when an application’s documentation says “open the required ports” without saying which, or when you are reviewing a rule set and need to know whether port 5900 belongs on the public internet. Every entry includes a plain-language security note, and where the port is defined by a standards document the relevant RFC is cited.

What You Can Search

  • By port number — type 3389 and get Remote Desktop Protocol, TCP, plus the reason it should never face the internet directly.
  • By service or keyword — type postgres, kerberos, or ldap to find the ports a service uses, including its secondary and alternate ports.
  • By protocol — filter to TCP only, UDP only, or entries registered for both.
  • By category — Database, Web, File Transfer, Mail, Monitoring, Streaming, Proxy, Time, Windows, Messaging, VoIP, and more.

Search an unassigned number and you still get an answer: the tool confirms the port is within the valid 0–65535 range and tells you it is unregistered, which is itself useful information when you are triaging unexpected traffic. Results are paginated, and the page size is adjustable for when you want to scan a whole category at once.

The Three Port Ranges

RangeNameWho assigns itTypical use
0–1023Well-known / system portsIANAStandard services: HTTP, SSH, DNS, SMTP
1024–49151Registered / user portsIANA on requestVendor applications: databases, management agents
49152–65535Dynamic / ephemeral portsNobody — allocated by the OSThe client side of outbound connections

On Unix-like systems, binding to a port below 1024 traditionally requires root or an explicit capability such as CAP_NET_BIND_SERVICE. That is why web servers historically started as root and dropped privileges, and why containerised applications usually listen on 8080 internally and are published on 443 by the platform. The ephemeral range is also why a firewall rule that only permits inbound 443 still works for outbound traffic — the reply arrives on a high-numbered port belonging to an established connection, which stateful firewalls track separately.

Ports You Will Look Up Most Often

PortProtocolServiceExposure guidance
22TCPSSHKey-only auth, restrict by source address
25 / 587 / 465TCPSMTP / submission / SMTPS25 for server-to-server, 587 for clients
53TCP + UDPDNSNever run an open recursive resolver
80 / 443TCPHTTP / HTTPSRedirect 80 to 443; the only pair most hosts need public
123UDPNTPRestrict monlist-style queries; classic amplification vector
445TCPSMBInternal only, always — the EternalBlue and ransomware path
1433 / 3306 / 5432TCPSQL Server / MySQL / PostgreSQLBind to localhost or a private subnet, never 0.0.0.0
3389TCPRDPBehind a VPN or gateway; the most brute-forced port on the internet
5900TCPVNCUnencrypted by default; tunnel over SSH
6379TCPRedisHistorically unauthenticated; bind to localhost

How to Use a Port Lookup in Practice

  1. Triaging unexpected traffic. Take the destination port from the firewall log and search it here. Knowing that 1900/UDP is SSDP, not an attack, saves an incident.
  2. Reviewing firewall rules. Filter by category — Database, for instance — and check whether any of those ports appear in a rule with a public source. Database ports on the internet are the single most common finding in an external assessment.
  3. Documenting an application. Search by service name to build the full list of ports an application needs, including the secondary ones that only appear under load or during clustering.
  4. Interpreting a port scan. Map each open port to a service, then ask whether that service is meant to be reachable from where the scan ran.

A caveat that matters: a registration tells you what a port is supposed to carry, not what it is carrying. Malware routinely uses 443 to blend into normal traffic, and administrators routinely move SSH to 2222. Confirm with the running process — ss -tulpn on Linux, netstat -ano on Windows — before drawing conclusions.

TCP or UDP?

TCP is connection-oriented: a three-way handshake, ordered delivery, retransmission of anything lost. UDP sends datagrams with none of that, which makes it faster and lighter but places reliability on the application. Services that need every byte in order — HTTP, SSH, SMTP, database protocols — use TCP. Services that value latency over completeness, or that fit in a single datagram, use UDP: DNS queries, NTP, SNMP, syslog, most real-time media.

Many services are registered for both, which is why the protocol column in this database has three values. DNS is the classic example: queries go over UDP/53, but zone transfers and responses too large for a datagram fall back to TCP/53. Block TCP/53 in the belief that DNS is “a UDP protocol” and DNSSEC-signed responses start failing intermittently. QUIC has since moved a great deal of web traffic onto UDP/443 as well, so a firewall that only permits TCP/443 will silently push clients back to TCP HTTP/2.

Frequently Asked Questions

How many ports are there in total?

65,536 per protocol, numbered 0 through 65535, because the port field in the TCP and UDP headers is 16 bits. This reference covers the 5,985 that have a registered or well-documented service associated with them.

What does it mean if a port is not in the database?

It is unregistered — typically an ephemeral port allocated by an operating system for an outbound connection, or a port an application picked for itself. The tool confirms the number is valid and tells you no assignment exists, which is the correct answer rather than an error.

Which ports are safe to expose to the internet?

As a default, 80 and 443 and nothing else. Everything else should sit behind a VPN, a bastion host, or a source-address restriction. Management protocols — SSH, RDP, VNC, SMB, WinRM — and every database port belong on a private network. Each entry here carries a security note reflecting that.

Can I run a service on a non-standard port?

Yes, and it is common. Moving SSH off 22 reduces automated scanning noise, though it is obscurity rather than security — a targeted scan finds it in seconds. The real controls are authentication, source restriction, and rate limiting.

Why do some entries list both TCP and UDP?

Because IANA registers a service name against a port number for both protocols even when only one is used in practice, and because some protocols genuinely use both — DNS on 53 being the standard example.

What is the difference between a closed port and a filtered port?

A closed port answers — TCP sends a RST — telling the scanner the host is up but nothing is listening. A filtered port produces no answer at all, because a firewall dropped the packet. Scanners report these differently, and the distinction tells you whether a firewall is in the path.

How do I see which ports are open on my own machine?

On Linux, ss -tulpn lists listening sockets with the owning process. On Windows, netstat -ano plus Task Manager maps a PID to a program. On macOS, lsof -iTCP -sTCP:LISTEN -n -P. Cross-reference the results here to confirm each listener is one you intended.

What else should I check alongside ports?

DNS records for the host — our DNS lookup tool resolves them — and the network boundaries themselves. The subnet calculator helps you define the source ranges a rule should permit, and the firewall rule simulator tests whether a rule set actually blocks what you think it does.

What Is a Port Reference Guide

A port reference provides a comprehensive lookup of TCP and UDP port numbers, their assigned protocols, and their security implications. Ports are logical endpoints that operating systems use to direct network traffic to the correct application—like apartment numbers in a building. Understanding which services run on which ports is fundamental to network administration, firewall configuration, and security operations.

The Internet Assigned Numbers Authority (IANA) maintains the official port registry, dividing the 65,535 available ports per protocol into three ranges: Well-Known Ports (0-1023), Registered Ports (1024-49151), and Dynamic/Ephemeral Ports (49152-65535). Security professionals need to know these assignments to interpret network scans, write firewall rules, and identify potentially malicious traffic.

How Network Ports Work

When a client connects to a server, it specifies both an IP address and a port number. The server listens on a specific port for incoming connections. The combination of IP address + port number + protocol (TCP or UDP) uniquely identifies a network service.

PortProtocolServiceSecurity Notes
22TCPSSHSecure remote access; target for brute-force attacks
25TCPSMTPEmail relay; often blocked to prevent spam
53TCP/UDPDNSName resolution; DNS tunneling risk
80TCPHTTPUnencrypted web traffic
443TCPHTTPSEncrypted web traffic (TLS)
445TCPSMBFile sharing; historically vulnerable (WannaCry)
1433TCPMSSQLDatabase; should never be internet-exposed
3306TCPMySQLDatabase; restrict to localhost or VPN
3389TCPRDPRemote desktop; major attack target
5432TCPPostgreSQLDatabase; restrict access carefully
8080TCPHTTP AltCommon for proxies and dev servers
8443TCPHTTPS AltCommon for management interfaces

Port states in scanning:

  • Open: A service is actively listening and accepting connections
  • Closed: The port is accessible but no service is listening
  • Filtered: A firewall is blocking the port; no response received

Common Use Cases

  • Firewall rule creation: Know which ports to allow/block based on required services
  • Network scanning interpretation: Understand what services Nmap or Nessus results indicate
  • Incident investigation: Identify unusual port activity that may indicate compromise or data exfiltration
  • Server hardening: Determine which ports should be open and close everything else
  • Application deployment: Configure the correct ports for web servers, databases, and APIs

Best Practices

  1. Follow the principle of least privilege — Only open ports that are required for business operations
  2. Never expose database ports to the internet — MySQL (3306), PostgreSQL (5432), and MSSQL (1433) should be accessible only from application servers
  3. Use non-standard ports cautiously — Moving SSH from 22 to another port reduces noise but is not a security control; attackers scan all ports
  4. Monitor for unauthorized listeners — Regularly audit which processes are listening on which ports with tools like netstat or ss
  5. Block outbound ports by default — Egress filtering prevents malware from establishing command-and-control connections on non-standard ports

Frequently Asked Questions

What are network ports and why do they matter?+

Network ports are logical endpoints in TCP/IP networking, numbered 0-65535, that enable multiple services to operate on a single IP address. Ports direct traffic to specific applications: web servers use 80/443, email uses 25/587/993, SSH uses 22. Understanding ports is essential for firewall configuration, troubleshooting connectivity, and securing network services properly.

What is the difference between TCP and UDP ports?+

TCP (Transmission Control Protocol) provides reliable, connection-oriented communication with error checking and guaranteed delivery. UDP (User Datagram Protocol) offers connectionless, faster communication without delivery guarantees. TCP suits applications needing reliability (HTTP, SSH, email); UDP suits speed-critical services (DNS, streaming, gaming). Many services support both protocols on the same port.

What are well-known ports vs registered ports?+

Well-known ports (0-1023) are reserved for common services by IANA and require administrator privileges. Examples: 22 (SSH), 80 (HTTP), 443 (HTTPS). Registered ports (1024-49151) are assigned to specific applications by IANA but don't require privileges. Dynamic/private ports (49152-65535) are temporary ports for client connections. Always verify port assignments before deployment.

Which ports should I block on my firewall?+

Block inbound connections on all ports except those explicitly required for services. High-risk ports include: 23 (Telnet), 135-139/445 (SMB/NetBIOS), 3389 (RDP), 5900 (VNC). Allow only: 80/443 for web, 22 for SSH (change default), required application ports. Implement egress filtering too, blocking unexpected outbound connections to prevent data exfiltration and command-and-control communication.

What are common ports used by attackers?+

Attackers frequently target: 22 (SSH brute force), 23 (Telnet default credentials), 80/443 (web vulnerabilities), 3389 (RDP attacks), 445 (SMB exploits like WannaCry), 21 (FTP), 1433/3306 (databases). They also scan high ports for backdoors and use non-standard ports to evade detection. Monitor all open ports, not just common targets, with intrusion detection systems.

How do I find what ports are open on my system?+

Windows: Run "netstat -ano" or "Get-NetTCPConnection" in PowerShell. Linux/Mac: Run "netstat -tuln", "ss -tuln", or "lsof -i". External scanning: Use nmap ("nmap -sT [IP]") or online port scanners. Always scan from both internal and external perspectives to understand exposure. Document all open ports and their purposes for security baseline and compliance.

Should I change default ports for security?+

Changing default ports provides minimal security through obscurity. It reduces automated scanning but determined attackers will discover services regardless. Better strategies: disable unnecessary services, use strong authentication, keep software updated, implement network segmentation, and monitor traffic. If changing ports, document changes carefully and ensure monitoring tools reflect new configurations to avoid blind spots.

What is port forwarding and when should I use it?+

Port forwarding redirects external traffic to internal network devices, enabling services like remote desktop, web servers, or game servers behind NAT. Configure on routers to map external ports to internal IP:port combinations. Use cautiously: each forwarded port increases attack surface. Prefer VPN access over port forwarding. If necessary, use non-standard ports, strong authentication, and monitor logs regularly.

Which ports should I block on my firewall?+

Block unused ports and known-vulnerable services like Telnet (23), NetBIOS (135-139, 445), and RDP (3389) unless needed. Allow only essential services through specific ports. Implement default-deny policies where all traffic is blocked except explicitly allowed connections.

What is the difference between TCP and UDP ports?+

TCP (Transmission Control Protocol) is connection-oriented with guaranteed delivery, ordering, and error checking - ideal for web, email, and file transfers. UDP (User Datagram Protocol) is connectionless with no delivery guarantees but lower latency - used for streaming, gaming, and DNS.

This tool is provided for informational and educational purposes only. All processing happens in your browser — no data is sent to or stored on our servers. While we strive for accuracy, we make no warranties about the completeness or reliability of results.